PROJECT SUMMARY Homologous recombination (HR) deficiency (HRD), particularly from biallelic mutational loss of BRCA1/BRCA2/ATM (BRCA/ATM), is significantly enriched in men with metastatic castration-resistant prostate cancer (mCRPC). Such patients have multiple FDA-approved systemic life-prolonging therapies to choose from, including abiraterone, enzalutamide, and taxane chemotherapies, as well as the possibility of poly(ADP- ribose) polymerase (PARP) inhibitor therapy which currently remains investigational. A few recent studies suggest that patients with germline and/or somatic HRD mutations may respond better (and for longer durations of time) to novel hormonal therapies than their HRD-negative counterparts. These studies suggest that in addition to PARP inhibition, potent AR suppression is also ?synthetic lethal? with HRD in mCRPC. However, the genetic/genomic determinants of HRD and their role in treatment selection remain unknown. We propose a resource-driven, patient-centered study to determine the genetic/genomic drivers of HRD predicting ?deep? response to abiraterone and enzalutamide. We hypothesize that mCRPC patients can be categorized into three groups according to HRD status defined by deleterious mutations in HRD genes: 1) germline/somatic HRD; 2) somatic-only HRD; 3) negative HRD; and that these groups are molecularly distinct, and have different clinical implications as predictive markers of response to AR-targeting therapies and taxane chemotherapies. To address the overall hypothesis, it is necessary to establish clinical and tumor/normal specimen cohorts that enable detailed molecular and clinical characterization of HRD in men with mCRPC. In Specific Aim 1, we will seek to ascertain the HRD mutations status, both somatic and germline, in three existing advanced/lethal prostate cancer cohorts enriched for HRD using blood-based assays. In Specific Aim 2, we will determine the association of HRD status defined by blood-based assays with treatment response to first-line AR-directed therapy (abiraterone/enzalutamide) and taxane chemotherapies in mCRPC patients by comparing treatment outcomes of men in these three groups. In Specific Aim 3, we seek to determine the expression correlates of HRD status defined by blood-based assays and further ascertained by tissue-based assays, by performing RNA-Seq in surgical specimens from men with lethal prostate cancer with: 1) germline/somatic HRD; 2) somatic-only HRD; and 3) negative HRD. The proposed work addresses an unmet need due to focus on liquid biopsy markers in a vulnerable patient population facing difficult treatment decisions. Also, our effort in defining the clinical and functional implications of HRD status in established cohorts of mCRPC patients will directly lead to treatment selection strategies to improve clinical management as well as patient selection strategies for clinical trials.